[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: default color of bar plots
From: |
Ben Abbott |
Subject: |
Re: default color of bar plots |
Date: |
Fri, 18 Nov 2011 18:53:53 -0500 |
On Nov 18, 2011, at 6:08 PM, Ben Abbott wrote:
>
> On Nov 18, 2011, at 3:21 PM, John W. Eaton wrote:
>
>> The changeset
>>
>> http://hg.savannah.gnu.org/hgweb/octave/rev/e36c2f4ea8f5
>>
>> caused the default color of bar plots to go from blue to a light green
>> color.
>>
>> I could undo the changeset, but I suspect that is not the correct fix.
>>
>> Is Octave handling clim, cdata, and cdatamapping correctly (as in
>> compatible with Matlab) for patch graphics objects?
>>
>> If not, what is supposed to be happening. It is my understanding that
>> patch objects are supposed to be colored according to the current
>> colormap, but I'm not sure what properties affect that behavior, or
>> how they are supposed to work. Some help with this would be much
>> appreciated.
>>
>> jwe
>
>
> I tried bar(1), which produces a blue bar for ML R2011b and an green one for
> Octave (tip: 99cd136079f7).
>
> For Matlab I see ...
>
> CData = [1;1;1;1]
> CDataMapping = 'scaled'
> FaceColor = 'flat'
>
> With clim = [1 2] for the axes.
>
> and for Octave ...
>
> cdata = 0
> cdatamapping = 'scaled'
> facecolor = 'flat'
>
> With clim = [-1 1] for the axes.
>
> The climmode for both Octave and Matlab is set to "auto".
>
> Ben
I tried a related example to try to get some more insight.
for c = 0:1
figure (1)
clf
patch ([0 0 1 1], [0 1 1 0], c)
clim = get (gca, 'clim')
end
For both Octave and ML, clim = c + [-1 1]
Perhaps Matlab caxis works differently for scalar cdata and vertex cdata?
Ben